Process
How to reduce self-selection bias in 6 steps
From defining the population to reporting who is missing
A six-step process for reducing self-selection bias in surveys and customer interviews: define the population, invite from the full list, make answering easy, follow up with non-responders, compare respondents with the list, then weight or report the gap. It suits researchers, product managers and customer success teams running feedback studies. Koji supports the invite and follow-up steps with personal interview links from a CSV and outbound phone interviews.
Where this appears
What's in this process
- Define who you need to hear from: Write the population down, such as everyone who cancelled in the last 30 days.
- Invite from the full list: Personally invite everyone, or a random sample, instead of posting an open link.
- Lower the effort to answer: Short, self-paced and mobile friendly, with a choice of voice or text.
- Follow up with the quiet ones: A reminder or a different channel, like a phone call, reaches people who ignored the first ask.
- Compare respondents with the list: Check plan, tenure, usage and segment to see who is over or under represented.
- Weight or report the gap: Weight results for missing groups, or say plainly the findings don't cover them.
Use this image
Free to use in your docs, decks and posts under CC BY 4.0, as long as you credit Koji with a link to this page. To embed it, paste this into your page:
html
<a href="https://www.koji.so/visuals/reduce-self-selection-bias-process"><img src="https://www.koji.so/visuals/reduce-self-selection-bias-process/reduce-self-selection-bias-process.png" alt="Six-step process to reduce self-selection bias, from defining the population to reporting who is missing" width="1080" height="1080" /></a>
<p>Source: <a href="https://www.koji.so/visuals/reduce-self-selection-bias-process">How to reduce self-selection bias in 6 steps</a>, Koji</p>
